December 2008
Typing or even twirling a pencil may seem like simple tasks to many, but Yoky Matsuoka, associate professor in the University of Washington’s Computer Science & Engineering Department, knows these seemingly simple tasks can be vital to a person’s everyday activities and quality of life.
Her research to develop a realistic prosthetic hand combines the fields of neuroscience and robotics, jointly called neurobotics. Discover the possibilities such research holds in “Where Humans and Robots Connect,” part of the new Engineering Lecture Series on UWTV.
Matsuoka directs the UW Neurobotics Lab, where one of her projects is to develop an anatomically correct robotic hand. Once finished, this prosthetic hand will mirror a human hand with its detailed make-up, including bone-shaped structures and more than 30 motors that act as muscles.
“There is a reason why very dexterous prosthetic devices are not available,” said Matsuoka. “It’s very difficult to put everything in a light, easy-to-use format. It’s still too premature, still a cutting-edge research topic.”
Consider the seemingly simple task of typing. Your brain does the thinking and your hands instantaneously move.
|
“There are electrical currents in the brain that are signals,” explained Matsuoka. “We try to extract that signal, decode it, then take that output in a computer process to amplify it and put it back into a robot.”
Such a complicated process makes the human hand difficult to replicate realistically.
“The hand has a lot more joints. If you think of an arm, you just have to move your elbows, and realistically, you can only bend it in one movement. Or a shoulder, which can go up and down, side to side, is controlling just the four degrees of freedom,” explained Matsuoka. “But with hands, all those joints have to curl around at the same time in a coordinated fashion and that’s the hard part of it. We have to somehow be able to decode a lot more information to do the right things with the hand.”
Though this may seem like a daunting feat to accomplish, Matsuoka’s lab is not far from developing a fully usable prosthetic hand that can assist people who have lost use of their fine motor skills. Currently, Matsuoka is working to perfect highly dexterous activities on a robotic model, such as spinning a pen or crumpling paper. She anticipates the robotic hand to be adept at these tasks within five years, though duplicating them when mounted on a person may take longer.
Introducing an anatomically correct prosthetic hand onto the public marketplace may take some time.
“To get to the point where our robotic hand is ‘live’ enough to have people using it as a prosthetic will take about 10 to 15 years,” said Matsuoka. “It’s a gradual process, and between now and 20 years from now, the current prosthetic companies will be improving and incorporating features that we have developed, ” said Matsuoka.“Hopefully a lot of people using prosthetic devices 10 years from now will be using prosthetic devices related to what we have developed, but it may not necessarily be our prosthetic hand.”
So how exactly can a robotic hand work in sync with the desires that we process in our brains?
“With time, we will find ways to gather more sophisticated data from both noninvasive techniques, such as imaging and skin sensors, and invasive techniques, such as embedded electronics in the tissue and the nervous system,” Matsuoka said. “Currently, these processes are still in the stages of infancy, only providing a couple pieces of information at a time.”
Matsuoka’s initial motivation for delving into the field of robotics sprung from her desire to create a robotic tennis partner. Now, her motivation comes from other people.
“In my research, I try to do things which are forward-looking in terms of things that can help people 25 years from now. Of course, scientific discoveries have to be made before companies can make it into products and sell it,” said Matsuoka. “But I am forming a nonprofit for people who need help now.”
Matsuoka’s work is not limited to prosthetics.
“It can be any sort of augmentation, for their room or wheelchair, their wearable device, attachment to some daily toothbrush that they use, it can be a variety of different things enhancing the ability for disabled people to enjoy life, play sports or play music that they are not able to do otherwise,” said Matsuoka.
The burgeoning field of neurobotics shows great potential and is rapidly growing. The research conducted by Matsuoka’s lab not only advances the field of neuroscience, but can further other fields of interest. How intertwined various fields are may surprise many.
“For example, building a robotic device that can move around in a human way influences the animation figures you might see in a theater because they also need to move in a very natural, human-like way,” said Matsuoka. “All the understanding we make will provide advances in all those fields.”
As part of the new Engineering Lecture Series, “Where Humans and Robots Connect” and other programs in the series aim to engage the public in the research that could directly affect their lives.
“One of the things is to really generate excitement that we are indeed making progress in science and engineering,” said Matsuoka. “And that ends up affecting people who need technology and science to feel better and improve their quality of life. When the excitement is there, more people pay attention and allow the field to move even further and faster.”
Look for two more programs in this series, “Beyond Oil: Powering the Future” and “Back to Nature for the Next Technology Revolution,” to debut on UWTV soon.
For more information about Insider features, contact Erin Lodi at erinlodi@u.washington.edu.
